Abstract
The exchange of information among health professionals is a common practice among clinics, laboratories, and hospitals. Cloud-based clinical data exchange platforms enable valuable information to be available in real time and in a secure and private manner. The increasing availability of data in health information systems allows specialists to extract knowledge using pattern recognition techniques for the identification and prediction of risk situations that could lead to severe complications for a patient. Hence, this paper proposes the use of a neuro-fuzzy machine learning technique for predicting the most complex hypertensive disorder in pregnancy called HELLP syndrome. This classifier serves as an inference mechanism for cloud-based mobile applications, for effective monitoring through the analysis of symptoms presented by pregnant women. Results show that the proposed model achieves excellent results regarding several indicators, such as precision (0.685), recall (0.756), the F-measure (0.705), and the area under the receiver operating characteristic curve (0.829). This technique can accurately predict situations that could lead to the death of both a mother and fetus, at any location and time.